Image Representation
Image representation research focuses on developing effective methods to encode visual information for computer vision tasks. Current efforts concentrate on improving the quality and efficiency of these representations, exploring diverse approaches such as implicit neural representations (INRs), hyperbolic graph neural networks, and contrastive learning frameworks, often integrated with multimodal data like text or sensor information. These advancements are crucial for enhancing the performance of various applications, including image segmentation, object recognition, and medical image analysis, by enabling more robust and efficient processing of visual data. The development of more effective image representations directly impacts the accuracy, speed, and resource efficiency of numerous computer vision systems.
Papers
PooDLe: Pooled and dense self-supervised learning from naturalistic videos
Alex N. Wang, Christopher Hoang, Yuwen Xiong, Yann LeCun, Mengye Ren
Deep Learning-based Classification of Dementia using Image Representation of Subcortical Signals
Shivani Ranjan, Ayush Tripathi, Harshal Shende, Robin Badal, Amit Kumar, Pramod Yadav, Deepak Joshi, Lalan Kumar